Lawrence Berkeley National Laboratory

Lawrence Berkeley National Laboratory

Title
Conceptual design of a compact positron tomograph for prostate imaging

Permalink
https://escholarship.org/uc/item/39w571wf

Authors
Huber, J.S.
Derenzo, S.E.
Qi, J.
et al.

Publication Date
2000-11-04

Peer reviewed

eScholarship.org Powered by the California Digital Library

University of California

Accepted to IEEE Transactions on Nuclear Science
Conceptual Design of a Compact Positron
Tomograph for Prostate Imaging
J. S. Huber, Member, IEEE, S. E. Derenzo, Fellow, IEEE, J. Qi, Member, IEEE, W. W. Moses,
Senior Member, IEEE, R. H. Huesman, Senior Member, IEEE, and T. F. Budinger, Member, IEEE
Abstract—We present a conceptual design of a compact
positron tomograph for prostate imaging using a pair of
external curved detector banks, one placed above and one
below the patient. The lower detector bank is fixed below the
patient bed, and the top bank adjusts vertically for maximum
sensitivity and patient access. Each bank is composed of 4 0
conventional block detectors, forming two arcs (44 cm minor,
60 cm major axis) that are tilted to minimize attenuation and
positioned as close as possible to the patient to improve
sensitivity. The individual detectors are angled to point
towards the prostate to minimize resolution degradation in
that region. Inter-plane septa extend 5 cm beyond the
scintillator crystals to reduce random and scatter
backgrounds. A patient is not fully encircled by detector rings
in order to minimize cost, causing incomplete sampling due to
the side gaps. Monte Carlo simulation (including randoms and
scatter) demonstrates the feasibility of detecting and
differentiating partial and whole prostate tumors with a tumor
to background ratio of 2:1, utilizing the number of events that
should be achievable with a 6-minute scan after a 10 mCi
injection (e.g., carbon-11 choline or fluorine-18 fluorocholine).
INTRODUCTION
I.
E present a conceptual design of a compact positron
W tomograph optimized to image the prostate. This
instrument images radiopharmaceuticals that specifically
localize in the prostate to confirm the presence, absence or
progression of disease. It will have approximately 4 times
less detectors than a conventional whole-body positron
emission tomograph (PET), which will reduce the cost and
increase clinical availability.
Prostate cancer has a prevalence and diagnostic rate similar
to breast cancer, with 360,000 new cases diagnosed each year
and 2 million men affected by the disease. Four percent of
men over 50 will have clinically significant disease [1].
Currently, prostate cancer suspicion is typically based on an
elevated prostate-specific antigen (PSA) level or a suspicious
node found during a digital rectal exam. Serum PSA values
do not always correlate well with clinical diagnosis or
outcomes [2-4]. Palpation is subjective, insensitive and
inexact; more than half of all cancers detected today are not
Manuscript received November 3, 2000. This work was supported by
the Director, Office of Energy Research, Office of Health and
Environmental Research, Medical Applications and Biophysical Research
Division of the U.S. Department of Energy under contract No. DE-AC03-
76SF00098.
J. S. Huber, S. E. Derenzo, J. Qi, W. W. Moses, R. H. Huesman and T.
F. Budinger are with the Lawrence Berkeley National Laboratory, Mailstop
55-121, 1 Cyclotron Road, Berkeley, CA 94720 USA (telephone: 510-486-
6445, e-mail: jshuber@lbl.gov).
1
LBNL-47068
palpable. Treatment decision is based mainly on biopsy
confirmation of prostate cancer, but the diagnostic accuracy of
prostate biopsies is problematic. Treatment usually consists
of radical prostatectomy, external beam irradiation,
brachytherapy (interstitial implantation of radioactive seeds),
androgen ablation (hormone) therapy, or "watchful waiting".
In some cases, radiation therapy is prescribed to the pelvis
and/or prostate bed even without confirmation by biopsy or
imaging [5-7], despite the fact that irradiating a wide field in
the pelvis (when disease is only suspected to extend beyond
the known cancerous prostate) can lead to a higher incidence
of post-irradiation morbidity than with irradiation of the
prostate bed alone [8]. A major problem with prostatic cancer
therapy is the question of when to treat or whether to treat at
all. This is particularly problematic in the case of an
increased PSA level after a prostatectomy. A new imaging
technology for sensitive detection of early stage prostate
cancer is needed to confirm PSA and help guide treatment
decisions. In addition, a new method is needed to assess
response shortly after treatment intervention.
In order to help meet these needs, we are designing a
compact PET camera optimized to image the prostate.
Functional PET imaging will help detect malignant tumors
in the prostate and/or prostate bed to confirm an increased
PSA level, as well as possibly help determine tumor
"aggressiveness" based on metabolic uptake levels, in order
to help guide whether to treat suspected prostatic cancer.
Although not optimized to detect distant metastatic disease,
this compact PET camera should also image local spread
beyond the prostate bed to help guide treatment decisions
such as whether a narrow or wide irradiation treatment field is
needed.
Many groups have shown that PET imaging with
18
[ F]fluorodeoxyglucose (FDG) is not a good technique for
prostate cancer diagnosis, because FDG is not very prostate
specific (with a standardized uptake value of typically ~2)
primarily due to bladder accumulation. However, more
promising PET radiopharmaceuticals have recently
demonstrated outstanding results in the sensitive detection of
prostate cancer, inspiring a new interest in using PET for
prostate cancer imaging. Consistent with the evidence of
increased pool size of choline in prostate cancer [9, 10], Hara
and co-workers demonstrated prostate tumor uptake by
[11C]choline. They find that: [11C]choline clears the blood
faster than FDG; its uptake in prostate tumors is significantly
higher than in normal and surrounding tissues [11, 12],

Accepted to IEEE Transactions on Nuclear Science
Fig. 1: [11C]choline image of prostate cancer (a) before and (b) after
treatment. These grayscale images indicate a high (white) uptake in the
prostate cancer compared with a low (gray) uptake elsewhere. Images are
provided by Hara and co-workers [13].
providing excellent tumor/normal contrast; and bladder
accumulation is minimal if the correct time course is chosen
[13] which is a major advantage over FDG. Additionally,
they observe that [11C]choline PET is more sensitive for
detecting bone metastases than bone scintigraphy. Therefore,
[11C]choline is an attractive PET tracer for imaging primary
and metastatic tumors of the prostate and potentially for other
regions of the body [14-18]. Figure 1 shows a [11C]choline
image of prostate cancer before and after therapy,
demonstrating the ability to detect prostate carcinoma using
choline and possibly its analogs. Other
11
C bank is fixed below the patient bed, and the top bank moves
radiopharmaceuticals are also under investigation for prostate
cancer PET imaging, including [11C]acetate and
11
[ C]methionine.
18
F imaging has the advantage of a longer half-life,
increasing commercial viability since an on-site cyclotron
facility would not be necessary (as it would for 11C imaging).
There are several 18F radiopharmaceuticals currently under
investigation for prostate cancer imaging, including
[18F]fluorocholine (FCH) [19, 20]. PET images using
[18F]fluorocholine demonstrate high standardized uptake
values (e.g., SUV of 8), indicating that FCH is well
localized in the prostate cancer and can be imaged with good
resolution if a short scan time (~5 minutes) is used.
Therefore, [18F]fluorocholine is also an attractive PET tracer
for imaging primary and metastatic tumors of the prostate.
2
LBNL-47068
(a)
Prostate
(b)
Upper Detector “Ring”
Adjustment Direction
Fig. 2: Proposed positron tomograph for prostate imaging (shielding not
shown). (a) Drawing of a transaxial view through prostate, showing the
patient centered between two detector banks. The individual detector
modules are angled to point towards the prostate. (b) Drawing of the
sagittal view. The bottom arc is fixed below the patient bed, whereas the
top arc adjusts vertically for patient access and compactness. Both detector
banks are tilted and positioned as close as possible to the prostate, which
improves sensitivity and minimizes attenuation.
II. CAMERA DESIGN
These new prostate tracers motivate us to design a low
cost PET camera optimized to image the prostate.
Coincidence imaging of positron emitters is achieved using a
pair of external curved detector banks, one placed above and
one below the patient. Fig. 2 shows the transaxial and
sagittal views of the proposed camera design. The bottom
upward for patient access and downward for maximum
sensitivity. Each bank consists of 40 conventional block
detectors such as those in the Siemens/CTI ECAT HR+ (8 x
8 array of 4.5 mm x 4.5 mm x 30 mm BGO crystals),
forming two arcs with a minor axis of 44 cm and major axis
of 60 cm. The proposed prostate camera has one-half the axial
coverage and uses about one-fourth the number of detectors as
in a conventional PET system. However, since the average
distance to the detectors is approximately one-half that of a
conventional 2D PET system, the solid angle for a central
source is approximately double.
Individual detector modules are angled to point towards
the camera center near the prostate location in order to reduce
penetration effects (in the detector) for annihilation photons
originating in the prostate, which is an unique feature for a
non-circular camera geometry. Annihilation photons from
other parts of the field of view (FOV) will suffer increased

Accepted to IEEE Transactions on Nuclear Science
Inter-plane
Septa
Lead Shielding
Fig. 3: Cut-out side view of the proposed positron tomograph, showing the
shielding and inter-plane septa.
penetration effects, but these FOV regions are less important.
Both detector banks are tilted to image the prostate while
minimizing attenuation (i.e., above the buttocks and below
the stomach), but the gantry can also allow zero tilt for thin
patients. The patient is not fully encircled in 2D, which will
be discussed further in Section III.
Standard block detectors with thick gamma ray detectors
(i.e., 3 attenuation lengths) are needed for good detection
efficiency. Narrow detector elements (i.e., 4.5 x 4.5 x 30 mm
BGO crystals) are desired to achieve good spatial resolution.
The 511 keV photons of primary interest originate from the
prostate which is roughly a small (e.g., 3 cm diameter)
spherical source with a known central location at a reasonable
distance from detector modules, so expensive detector
modules with depth of interaction capability are not
necessary. A two ring camera is more practical for patient
positioning. Therefore, our proposed camera design attempts
to maximize performance while minimizing cost, in order to
produce a PET camera with nearly twice the detection
efficiency as a conventional 2D PET camera with
approximately one-quarter the detector costs.
The camera design also includes shielding, as shown in
Fig. 3. Inter-plane septa extend 5 cm beyond the scintillator
crystals to reduce background events from random
100
80
60
40
Angle (Degree)
20
0 an iterative maximum likelihood algorithm [22]. Self-
−20
−40
−60
−80
−100
−300 −200 −100 0 100 200 300
Radial Distance (mm)
Fig. 4: Incomplete sampling for the field of view due to side gaps between
detector banks. Coincidences between detector modules within the same
bank and imaging plane are allowed, resulting in the ~triangular sampling
regions on the left and right. The field of view of the proposed camera is
an elliptical region, so the only missing samples are the “X” region in the
center of the sinogram.
3
LBNL-47068
−150
−100
−50
50
100
150
−200 −100 0 100 200
Fig. 5: Statistical noise-free reconstructed image of five point sources,
demonstrating the local impulse response function and sampling
sufficiency. Scale in millimeters.
coincidences and from photons that Compton scatter in the
patient. Lead shields are also used on the ends to reduce
activity from outside the field of view.
III. EXPECTED PERFORMANCE
A. Simulation
Our camera geometry provides a compact design (i.e., 44
cm minor axis between opposing detector faces) that adjusts
vertically to fit the patient. However, a patient of average size
will not be fully encircled by the detector rings, as mentioned
in Section II. Fig. 4 shows the resulting irregular and
incomplete sampling due to the side gaps. Similar irregular
and incomplete sampling patterns have been dealt with in our
positron emission mammography camera [21]. Despite this
incomplete sampling, we are able to reconstruct nearly artifact
free images in the region of interest by using an iterative
reconstruction as described below. This iterative
reconstruction algorithm can easily deal with the missing
data of the system and can also partially recover the
resolution loss (due to the crystal penetration effect) in the
off-center FOV region. The gap size will increase for larger
patients and this may impact the image quality, but we have
not yet explored the effects of variations in gap size.
Our camera design includes inter-plane septa, so the data
will be acquired in 2D. We use Monte Carlo data to simulate
the camera performance. Reconstructions are performed with
attenuation is simulated assuming a constant attenuation
coefficient of 0.0095/mm inside the body, then attenuation is
modeled in the reconstruction. Coincidences between detector
modules within the same bank are allowed.
Fig. 5 shows an image of five point sources that are
reconstructed without statistical noise to demonstrate how
irregular sampling affects the local response function. All
point sources are clearly visible. We observe increased radial
blurring near the outer surface of the patient as expected due
to penetration effects, but observe good resolution in the
region near the center (i.e., near the prostate). In short, we are
optimizing resolution at the center near the prostate at the
price of resolution at the edges, which implies that distant

Accepted to IEEE Transactions on Nuclear Science
metastases will suffer from some resolution degradation. Data
are generated by forward projection of 256x256 2 mm x 2
mm phantom voxels to simulate real data, and reconstruction
is done with 128x128 4.51 mm x 4.51 mm pixels, thus
causing some point source spread among adjacent pixels.
As we are planning to use iterative methods to reconstruct
images, the anticipated spatial resolution will depend on the
geometrical parameters used for reconstruction (e.g., the
voxel size and detector model) as well as the number of
events. Thus, it is difficult to derive meaningful estimates of
the expected spatial resolution from the test image in Fig. 5.
Therefore, we estimate the spatial resolution analytically with
the help of a simplifying assumption. That assumption is
that if two tomographs employ the same scintillation
material, the contribution of penetration to the spatial
resolution for a source at any position within either camera
depends only on the angle θ that the incoming gamma ray
makes to the axis of the detector crystals (see Fig. 6). Given
this assumption, we can use measured spatial resolutions [23]
to estimate the resolution for the proposed camera. Fig. 6(b)
(a)
θ differentiate partial and whole prostate tumors. Our
y preliminary rate calculations (described below) indicate that
x 745 kcounts per imaging plane will be achievable with a 6-
(b)
16
Radial Resolution (mm fwhm)
14 X-Direction (60 cm)
Y-Direction (44 cm)
12 Circular Camera (60 cm)
10
8 per slice is nearly 2 times higher than the HR (i.e., our
6
4 in a total system rate that is approximately the same as the
2 HR. Forty five minutes after injecting 15 mCi of FDG into
0
0 5 10 15
Distance from Tomograph Center (cm)
Fig. 6: (a) Proposed prostate PET camera drawing shown with a source
within the camera. The angle θ is defined to be that which an incoming
gamma ray makes to the axis of the detector crystal. (b) The anticipated
spatial resolution for the proposed prostate PET camera as a function of
distance from the center of the field of view along both the x- (square) and
y- (circle) axes, based on measured spatial resolutions with a circular
tomograph. Little degradation due to penetration artifacts is seen for objects
within 5 cm of the camera center, and 6 mm fwhm or better resolution is
obtained within 8 cm of the camera center. Plot also shows spatial
resolution for the reference circular PET camera as a function of distance
from the center of the FOV along the x- (triangle) axis.
4
LBNL-47068
shows the anticipated spatial resolution obtained in this
manner as a function of distance from the center of the field
of view (along both the x- and y- axes). Notice that little
degradation due to penetration artifacts is seen for objects
within 5 cm of the camera center, and that 6 mm fwhm or
better resolution is obtained within 8 cm of the camera
center. This region is easily large enough to contain the entire
prostate.
We also use Monte Carlo data to simulate the performance
of our proposed camera for imaging extended sources.
Extended “prostate” sources are reconstructed with statistical
noise representing 745 kcounts using a preliminary iterative
maximum likelihood algorithm. We assume a 2.5 cm
diameter spherical “prostate” (see fig. 7(a)) with a tumor to
background ratio of 2:1, a 30% scatter fraction and 20%
randoms fraction. The images are post-filtered with a
Gaussian function having a standard deviation of 1 pixel. The
spherical "prostate" is divided into quadrants. Fig. 7 (b)
shows a reconstructed image with a tumor in the upper left
quadrant of the spherical prostate. Fig. 7 (c) shows a
reconstructed image with a tumor in the right half of the
prostate. Fig 7 (d) shows a reconstructed image with a tumor
in the upper right and lower left quadrants. These "prostate"
tumors are clearly visible in all images, demonstrating that
we can reconstruct nearly artifact free images near the prostate
despite incomplete sampling and that the scanner can
minute scan after injecting a clinically acceptable (10 mCi)
patient dose of [11C]choline or [18F]fluorocholine. A relatively
short value of six minutes is chosen to demonstrate that good
images can be obtained even if the bladder fills rapidly. All
published [11C]choline or [18F]fluorocholine studies done to
date have used three minutes or longer (up to 30 minutes)
imaging times.
B. Estimated Rates
The average distance between detector banks in our
proposed camera is approximately one-half that of a
conventional PET system such as the HR, so our sensitivity
detectors will have nearly twice the rate). However, our
camera also has half the number of imaging planes, resulting
an average-sized patient (i.e., with ~10 mCi remaining in the
20 patient), we see a 50 kHz total coincidence rate using a HR
scanner. Therefore, our total system coincidence rate is
expected to be approximately 50 kHz, with a detector live
time of at least 80%.
For our proposed camera geometry, there are 166 total
cross planes. Assuming a 50 kHz total coincidence rate, this
implies a rate of 300 Hz per cross plane. Summing over 7
cross plane slices per imaging plane, this corresponds to a 2
kHz coincidence rate per imaging plane. Therefore, 745
kcounts per imaging plane will be achievable with a 6-minute

Accepted to IEEE Transactions on Nuclear Science
Fig. 7: Spherical "prostate" (2.5 cm diameter) is divided into quadrants, as
shown in (a). Reconstructed image with a tumor in the (b) upper left
quadrant, (c) right half, and (d) upper right and lower left quadrants of the
prostate, assuming a tumor to background ratio of 2:1, 30% scatter fraction
and 20% randoms fraction. All prostate tumors are clearly visible in all
images, demonstrating that we can reconstruct nearly artifact free images
near the prostate despite incomplete sampling and that the scanner can
differentiate partial and whole prostate tumors.
scan after injecting a clinically acceptable (10 mCi) patient
dose of [11C]choline or [18F]fluorocholine.
Rates are also computed using analytic estimates that were
developed in our research group [24] and fit to published
sensitivity and backgrounds of the HR tomograph [25]. They
assume a 10 mCi injection of [11C]choline or
[18F]fluorocholine uniformly distributed in a 70 kG subject
5
LBNL-47068
(torso 36 cm major axis, 17 cm minor axis), resulting in an
activity concentration of 0.15 µCi per ml. These calculated
rates are in agreement with patient data within a factor of 2.
IV. DATA CORRECTION
This design does not include hardware for obtaining data
to perform attenuation correction. The proposed design has
sufficient room to incorporate a single photon or positron
emitting source for transmission scans, but it would add to
the camera cost. However, many clinical whole-body PET
scans are done without an attenuation scan so it may not be
necessary for the detection of abnormal activity in the
prostate bed. A practical approach is to correct without
additional hardware using calculated attenuation coefficients
based on body contours (obtained from emission data) and an
uniform attenuation coefficient. Anatomical boundaries are
obtained from the outer edges of emission sinograms
acquired from transverse sections [26]. This method can be
generalized to three dimensions, and a uniform soft tissue
attenuation coefficient can be used to compute the attenuation
correction factors. This method assumes that the distribution
of attenuating medium is uniform, that its external boundary
is convex, that the attenuation coefficient for 511 keV
photons in soft tissue is uniform, and that inter-subject
variability is small. These assumptions should be valid for
the geometry and anatomy being imaged in this work.
Attenuation can also be estimated using X-ray CT images to
identify tissue types (bone, soft tissue, air, etc.), but this
requires a second imaging procedure.
Because of the unusual geometry of our proposed camera,
the sensitivity of each line of response depends on the
distance between the two detectors; the shorter the distance
is, the higher the sensitivity. Normalization will include
geometric correction (sensitivity changes caused by the septa
that are not modeled in the solid angle computation), block
effect correction, individual detector sensitivity correction,
and dead time correction. These factors will be obtained
using a scan of an uniform activity cylindrical source such as
used in conventional PET scanners. However, because of our
specialized geometry and lack of rotational symmetry, the
method for deriving the first two factors will be different
from that of circular systems and may require a longer scan
time to achieve good count statistics.
Random events will be estimated using the delayed
window technique. Instead of pre-subtracting histogrammed
data, delayed events will be saved in the list mode data
stream with a tag to distinguish them from prompt events.
Such information can be used to obtain a better estimate of
the mean of the random events [27]. For our statistical
reconstruction algorithm, the estimated randoms will be
included in the forward model in reconstruction. Anatomical
registration can be done using ultrasound or a point source.
V. DISCUSSION
This project is in its infancy, so there are many remaining
uncertainties such as: which radiopharmaceutical will be
used, the specific requirements for the radiopharmaceutical

Accepted to IEEE Transactions on Nuclear Science
and imaging procedure, and the accuracy that physicians
would require to perform effective diagnosis and evaluate
response to therapy. However, new PET radiopharmaceuticals
have recently demonstrated promising results in the sensitive
detection of prostate cancer. Hence, a low cost PET scanner
coupled with a 18F-based isotope could make PET diagnosis
of prostate cancer more clinically available. This PET camera
design is also likely to out-perform whole body PET for this
task, as its 2 times higher sensitivity will help for the short
scans that appear to be necessary with choline-like
radiopharmaceuticals due to bladder accumulation. We
believe that providing improved PET imaging performance
near the prostate bed at lower cost (than commercial PET) is
clinically significant, outweighing the trade-off of some
resolution degradation for distant metastases.
VI. CONCLUSION
Promising new PET tracers for prostate cancer, such as
[11C]choline and [18F]fluorocholine, motivate us to build a
low cost PET camera optimized for prostate imaging. Monte
Carlo simulations demonstrate the feasibility of detecting and
differentiating partial and whole prostate tumors with a tumor
to background ratio of 2:1, using a 6-minute scan after a 10
mCi injection. Thus, the proposed camera will image
prostate tumors with good spatial resolution and image
contrast at low cost relative to commercial PET scanners.
VII. ACKNOWLEDGMENT
We appreciate Dr. Hara from the International Medical
Center of Japan for providing the PET images presented in
Fig. 1 of this paper. This work was supported by the
Director, Office of Science, Office of Biological and
Environmental Research, Medical Science Division of the
U.S. Department of Energy under contract No. DE-AC03-
76SF00098. Reference to a company or product name does
not imply approval or recommendation by the University of
California or the U.S. Department of Energy to the exclusion
of others that may be suitable.
VIII. REFERENCES
[1] http://osp.nci.nih.gov/PRGReports/PPRGReport/toc.htm.
[2] H. I. Scher and L. W. Chung, “Bone metastases: improving the
therapeutic index,” Semin Oncol, vol. 21, pp. 630-56, 1994.
[3] H. I. Scher, M. Mazumdar, and W. K. Kelly, “Clinical trials in
relapsed prostate cancer: defining the target,” J Natl Cancer Inst, vol.
88, pp. 1623-34, 1996.
[4] M. A. Eisenberger and W. G. Nelson, “How much can we rely on the
level of prostate-specific antigen as an end point for evaluation of
clinical trials? A word of caution!,” J Natl Cancer Inst, vol. 88, pp.
779-81, 1996.
[5] S. E. Schild, W. W. Wong, G. L. Grado, et al., “Radiotherapy for
isolated increases in serum prostate-specific antigen levels after
radical prostatectomy,” Mayo Clin Proc, vol. 69, pp. 613-9, 1994.
[6] C. A. Perez and A. Eisbruch, “Role of post-radical prostatectomy
irradiation in carcinoma of the prostate,” Semin Rad Oncol, vol. 3, pp.
198, 1993.
[7] B. E. Shevlin, B. B. Mittal, W. N. Brand, et al., “The role of adjuvant
irradiation following primary prostatectomy, based on histopathologic
extent of tumor,” Int J Rad Oncol Biol Phys, vol. 16, pp. 1425-30,
1989.
[8] J. S. Horoszewicz, E. Kawinski, and G. P. Murphy, “Monoclonal
antibodies to a new antigenic marker in epithelial prostatic cells and
6
LBNL-47068
serum of prostatic cancer patients,” Anticancer Res, vol. 7, pp. 927-
35, 1987.
[9] J. Kurhanewicz, D. B. Vigneron, H. Hricak, et al., “Three-
dimensional H-1 MR spectroscopic imaging of the in situ human
prostate with high (0.24-0.7-cm3) spatial resolution,” Radiology, vol.
198, pp. 795-805, 1996.
[10] J. Kurhanewicz, D. B. Vigneron, R. G. Males, et al., “The Prostate:
Magnetic Resonance Imaging and Spectroscopy: Present and Future,”
in Radiologic Clinics of North America, H. Hricak and P. R. Carroll,
Eds. New York, New York: W. B. Saunders Co., 2000, pp. 115-138.
[11] T. Hara, N. Kosaka, N. Shinoura, et al., “PET imaging of brain tumor
with [methyl-C11]choline,” J Nucl Med, vol. 38, pp. 842-7, 1997.
[12] N. Shinoura, M. Nishijima, T. Hara, et al., “Brain tumors: detection
with C-11 choline PET,” Radiology, vol. 202, pp. 497-503, 1997.
[13] T. Hara, N. Kosaka, and H. Kishi, “PET imaging of prostate cancer
using carbon-11-choline,” J Nucl Med, vol. 39, pp. 990-5, 1998.
[14] T. Hara, N. Kosaka, T. Kondo, et al., “Imaging of brain tumor, lung
cancer, esophagus cancer, colon cancer, prostate cancer, and
bladder cancer with [C-11]choline,” J Nucl Med, vol. 38 (suppl), pp.
250P (abstract), 1997.
[15] T. Hara, K. Inagaki, N. Kosaka, et al., “Sensitive detection of
mediastinal lymph node metastasis of lung cancer with 11C-choline
PET,” J Nucl Med, vol. 41, pp. 1507-13, 2000.
[16] I. J. Jong, T. H. Que, J. Pruim, et al., “Imaging of bladder cancer
using carbon-11 choline positron emission tomography,” J. Nucl.
Med., vol. 41 (5 Suppl), pp. 74, 2000.
[17] J. Kotzerke, J. U. Prang, B. Neumaier, et al., “Carbon-11 choline
positron emission tomography (PET) of prostate cancer -- first clinical
experience,” J. Nucl. Med., vol. 41 (5 Suppl), pp. 74, 2000.
[18] J. Kotzerke, J. Prang, B. Neumaier, et al., “Experience with carbon-
11 choline positron emission tomography in prostate carcinoma,” Eur
J Nucl Med, vol. 27, pp. 1415-9, 2000.
[19] T. R. Degrado, R. E. Coleman, S. W. Baldwin, et al., “Fluorine-18
fluorocholine (FCH) as an oncological PET tracer: evaluation in
murine prostate cancer xenograft model,” J. Nucl. Med., vol. 41 (5
Suppl), pp. 231, 2000.
[20] T. R. Degrado, R. E. Coleman, S. Wang, et al., “Synthesis and
Evaluation of F18-labeled Choline as an Oncologic Tracer for
Positron Emission Tomography: Initial Findings in Prostate Cancer,”
Cancer Research, vol. 61(1), pp. 110-117, 2001.
[21] P. R. G. Virador, W. W. Moses, and R. H. Huesman, “Reconstruction
in PET cameras with irregular sampling and depth of interaction
capability,” IEEE Trans Nucl Sci, vol. NS-45, pp. 1225–30, 1998.
[22] J. Qi, R. M. Leahy, S. R. Cherry, et al., “High-resolution 3D Bayesian
image reconstruction using the microPET small-animal scanner,”
Phys Med Biol, vol. 43, pp. 1001-13, 1998.
[23] S. E. Derenzo, R. H. Huesman, J. L. Cahoon, et al., “Positron
tomograph with 600 BGO crystals and 2.6 mm resolution,” IEEE
Trans Nucl Sci, vol. NS-35, pp. 659-64, 1988.
[24] S. E. Derenzo, “Method for optimizing side shielding in positron
emission tomographs and for comparing detector materials,” J Nucl
Med, vol. 21, pp. 971-977, 1980.
[25] K. Wienhard, M. Dahlbom, L. Eriksson, et al., “The ECAT EXACT
HR: performance of a new high resolution positron scanner,” J Comp
Asst Tom, vol. 18, pp. 110-118, 1994.
[26] T. F. Budinger and G. T. Gullberg, “Three-dimensional reconstruction
in nuclear medicine emission imaging.,” IEEE Trans Nucl Sci,, vol.
NS-21, pp. 2-20, 1974.
[27] E. Mumcuoglu, R. M. Leahy, and S. R. Cherry, “Bayesian
reconstruction of PET images: methodology and performance
analysis,” Phys Med Bio, vol. 41, pp. 1777-1807, 1996.